Defoliation of Soybean Expressing Cry1Ac by Lepidopteran Pests

Lepidoptera, stink bugs, and weevils are important pests in soybean. For lepidopteran control, insecticides and seed treatments are used. As an alternative, Bt soybean was developed to control primary pests of Lepidoptera such as Rachiplusia nu (Guenée) (Noctuidae), Chrysodeixis includens (Walker) (Noctuidae), Anticarsia gemmatalis Hübner (Erebidae), Helicoverpa gelotopoeon (Dyar) (Noctuidae), and Crocidosema aporema (Walsingham) (Tortricidae). However, the use of transgenic plants, and the resulting reduction of insecticide against target pests, may allow other pest species to become more prevalent in agricultural environments. Soybean expressing Cry1Ac against different lepidopteran nontarget and target insect pests was evaluated, and its performance was compared with non-Bt soybean with seed treatment. The treatments were Bt soybean, non-Bt soybean with seed treatment (Fortenza® diamide insecticide, Syngenta, Buenos Aires, Argentina), and non-Bt soybean without seed treatment. Larvae of H. gelotopoeon, Spodoptera albula (Walker) (Noctuidae), Spodoptera cosmiodes (Walker) (Noctuidae), Spodoptera eridania (Stoll) (Noctuidae), and Spodoptera frugiperda (J. E. Smith) (Lep.: Noctuidae) were used. The plants of each treatment were infested with larvae of each species, and the percentage of leaf damage produced by each species was recorded. The results showed that Bt soybean provided control of H. gelotopoeon and had a suppressive effect on S. frugiperda and S. albula. However, S. eridania and S. cosmiodes were not susceptible to the Cry1Ac protein in MON 87701 × MON 89788 soybean when evaluated by greenhouse infestation. Considering the performance of each species using non-Bt soybean without seed treatment, S. eridania would represent a potential risk in soybean crops.


Introduction
Soybean (Glycine max (L.) Merrill) is an oilseed plant of the family Fabaceae that is widely planted in various countries worldwide [1]. Argentina is the third-major soybean producer in the world, covering an area of 19.2 million hectares [2,3]. Lepidoptera, stink bugs (Hem.: Pentatomidae), and weevils (Col.: Curculionidae) are important pests that inflict the major yield loss of this crop [1,[4][5][6][7][8]. For Lepidoptera in soybean, insecticides and seed treatments are used, but Bacillus thuringiensis (Berliner 1915) soybean (Bt soybean) is a new alternative method to control several  Helicoverpa gelotopoeon is another major pest of soybean. Larvae of this species cause damage in the vegetative and reproductive plant growth stages. In Tucumán and other provinces of Argentina, this species causes severe damage to soybean and chickpea (Cicer arietinum L. (Fabaceae)) crops and can be difficult to control with insecticides [50][51][52][53]. Some other species of Helicoverpa, such as H. armigera Hübner and H. zea, have also developed resistance to insecticides and Cry proteins [54][55][56][57][58][59].
Previous studies were conducted under field conditions considering the Helicoverpa genus in general [7,8,19]. This has led to much speculation, given the coexistence of H. gelotopoeon and H. armigera in soybean crops in northwestern Argentina [52]. In this genus, species can only be separated by morphological characters of the adults [50]. Thus, the study of H. gelotopoeon performance is important to know the real potential of defoliation in soybean crops.
The rapid adoption of Bt soybean in Argentina [2] and other countries in South America [61] has increased the need to know the behavior of nontarget and target pests of Bt soybean. The objective of this study was to evaluate the performance of soybean expressing Cry1Ac against different lepidopteran nontarget and target insect pests and to compare its performance with that of a non-Bt soybean with seed treatment. We addressed this study by evaluating the defoliation produced by different lepidopteran pests recorded at eight days after infestation.

Larval Collection
Larvae of H. gelotopoeon, S. albula, S. cosmiodes, S. eridania, and S. frugiperda were collected from January to March 2014 in commercial soybean fields in La Cocha (Dpto La Cocha) and Overo Pozo (Dpto Cruz Alta) counties in Tucumán province (Argentina). A minimum of 250 larvae (from 3rd to 6th instars) of each species were collected using a vertical beat sheet [62]. Then, each larva was placed in a glass tube (12 cm H and 1.5 cm D) with leaves of soybean and transported to our laboratories at the Estación Experimental Agroindustrial Obispo Colombres (EEAOC). The collected larvae were placed in growth chambers under controlled conditions (27 ± 2 • C, 70 ± 5% RH, 14:10 h L:D) until adult emergence. Late larval instars and adults were examined using morphological characters both to confirm the identity of species [47,49,63,64] and to establish pure cultures for each species in the laboratory. Voucher specimens for each species were deposited in the collection of Sección Zoología Agrícola, (EEAOC) Tucumán, Argentina.

Insect Rearing
Each species colony was maintained in the same chamber under identically controlled conditions at 27 ± 2 • C, 70 ± 5% RH, 14:10 h L:D. Colonies were reared according to the methodology described by Murúa et al. [65] and Herrero et al. [53]. Twenty-five pairs per cage and per species (25 females and 25 males) (N = 6) were used. Adults were maintained in cylindrical oviposition cages made out of plastic mesh (20 cm high and 15 cm in diameter) lined with polyethylene bags as an oviposition substrate. For aeration, both ends of the cage were covered with a nylon cloth. The food was provided via a cotton plug saturated with a mixture of honey and water (1:1 v/v) which was replaced every day. Cages were checked daily for oviposition and adult mortality. Eggs were collected daily with a moistened brush and deposited in Petri dishes lined with moistened filter paper. Once emerged, neonate larvae were placed in 20-cm diameter, 800-mL containers with an artificial larval diet that included bean flour (Grandiet ® , Buenos Aires, Argentina), wheat germ (Grandiet ® , Buenos Aires, Argentina), brewer's yeast (Calsa ® , Tucumán, Argentina), vitamin C (Anedra ® , Buenos Aires, Argentina), sorbic acid (Anedra ® , Buenos Aires, Argentina), vitamin supplement amino acids (Ruminal ® , Buenos Aires, Argentina), and methylparaben (Todo Droga ® , Córdoba, Argentina). Diet was replaced every five days. As larvae pupated, pupae were sexed and placed in cup containers with moistened filter paper until adult's emergence. Adults were used to initiate a new generation. After establishing a colony for each species, individuals from the 2nd generation (F2) were used for the evaluation of soybean Cry1Ac against different pest species in a greenhouse.

Greenhouse Studies
Soybean seeds of maturity group 7 were used (recommended for cultivation in northwestern Argentina). The treatments evaluated were Bt soybean Cry1Ac (T1), non-Bt soybean plus seed treatment (T2) (Fortenza ® diamide insecticide, Syngenta, Argentina) according to the dose/rate recommended by the company (36 g a.i./100 Kg/seed), and non-Bt soybean without seed treatment as an untreated control (T3). The three treatments were planted in different pots (15 cm D, 600 mL) using sterilized soil. One seed per pot was planted and each plant was labeled to distinguish both the species and treatment. The plants were maintained under greenhouse conditions under ambient lighting at approximately 33 ± 4 • C, 80 ± 10% RH, 14:10 h L:D.
The experimental design was completely randomized with three replicates per treatment, where each replicate consisted of 20 soybean plants. A total of 60 plants for each pest species (5) per treatment (3) were evaluated, resulting in 900 plants.
Each plant for each treatment was inoculated at the V1 stage [66] (approximately 14 days after planting) with 10 larvae (L1) of a pest species. The evaluations of defoliation produced by each species were recorded at 8 days after inoculation (DAI), according the visual estimation of defoliation scale described by Kogan and Turnipseed [67].
Expression of the Cry1Ac protein in the soybean plants was confirmed using qualitative ELISA Quickstix lateral flow detection strips (Envirologix, Portland, ME, USA).

Data Analysis
To meet parametric assumptions, percentage data on defoliation damage were transformed to arcsine square root prior to analysis [68]; nevertheless, untransformed means (±SE) are shown in the figure. The percentage of defoliation damage produced by each species in the different treatments and the performance of these species using non-Bt soybean without seed treatment were analyzed using a one-way ANOVA, and means were separated using Tukey's tests (p < 0.05) with InfoStat [69].

Results
All results are shown in Figure 1. Significant differences were found in the average defoliation damage produced by H. gelotopoeon among different treatments (F = 64.3; df = 2,177; p < 0.0001). Bt soybean (T1) and non-Bt soybean plus Fortenza ® (T2) presented similar values, but significant differences were found when they were compared with the control (T3). The average defoliation damage produced by S. albula was lower in the Bt soybean and non-Bt soybean plus Fortenza ® treatments compared to the control (Figure 1) (F = 111.0; df = 2,177; p < 0.0001). For S. cosmiodes, lower feeding damage was recorded in the non-Bt soybean plus Fortenza ® treatment, compared to the Bt soybean and control plants (F = 76.0; df = 2,176; p < 0.0001). The best control of S. eridania was with the non-Bt soybean plus Fortenza ® treatment, registering a lower percentage of defoliation compared to Bt soybean and the control plants (F = 269.1; df = 2,177; p < 0.0001). Finally, significant differences were found in the percentage of defoliation produced by S. frugiperda in the different treatments (F = 94.7; df = 2,177; p < 0.0001). The lowest defoliation was obtained with non-Bt soybean plus Fortenza ® (0.57 ± 0.12), whereas the control plants showed the highest defoliation.

Discussion
This study was conducted to characterize the efficacy of soybean expressing Cry1Ac against H. gelotopoeon, S. albula, S. cosmiodes, S. eridania, and S. frugiperda and compare its performance with that of a non-Bt soybean with seed treatment by assessing defoliation (Figure 1). On the other hand, the performance analysis made among the species using non-Bt soybean without seed treatment showed which of these species would represent a potential risk for soybean crops. For both analyses, the defoliation and potential risk were calculated considering the identity for each species in greenhouse conditions (Table 1).
Our results showed that Bt soybean provided variable control of some pests evaluated. This new technology had the best control of H. gelotopoeon and a suppressive effect on S. frugiperda and S. albula.
Non-Bt soybean with seed treatment provided good control of the five species. Nevertheless, this control will be only for the early plant growth stage, when plants have more insecticide active ingredient in their system. S. cosmiodes, S. eridania, S. frugiperda, and S. albula feeding resulted in the lowest percentage of defoliation compared with the other treatments.
In the case of S. albula, the defoliation recorded in T1 was similar to that recorded in T2 and the differences were not significant.
These results are consistent with field studies evaluating Bt soybean in Brazil and Argentina. A study of diversity, composition, and population dynamics of arthropods in non-Bt soybean and Bt soybean showed that A. gemmatalis, C. includens, C. virescens, and S. frugiperda were significantly controlled by Bt soybean. However, other species of the Spodoptera complex were not controlled by this technology [70]. Other studies reported that S. cosmiodes was the most abundant species attacking Bt soybean [71] and that development and reproduction of this species were not affected by the Cry1Ac protein [20]. A field study showed that the lepidopteran pests recorded during the soybean cycle were R. nu, A. gemmatalis, H. gelotopoeon, S. frugiperda, and S. cosmioides [72] and the population abundance of the last two species was similar in Bt and non-Bt soybean. In northwestern Argentina, a trial conducted to evaluate Bt soybean against different insect pest and natural predators found that this new technology provided good control of target pests such as A. gemmatalis, C. includens, and R. nu. The levels of leaf damage observed with Bt soybean were lowest and significantly different to those obtained with non-Bt soybean [4]. On the other hand, bioassays and greenhouse studies found that S. cosmiodes, S. eridania, and S. frugiperda exhibited low to no susceptibility to MON 87701 × MON 89788 soybean containing the protein Cry1Ac, and these species showed higher tolerance to the Cry1Ac protein than other Lepidoptera species, such as C. includens, Percentage of defoliation produced by Helicoverpa gelotopoeon, Spodoptera albula, Spodoptera cosmiodes, Spodoptera eridania, and Spodoptera frugiperda (Lepidoptera: Noctuidae) larvae in different soybean treatments (T1: Bt soybean, T2: non-Bt soybean with seed treatment, T3: non-Bt soybean without seed treatment). Means ± SE within species accompanied by different letters indicate significant differences (Tukey test, p < 0.05).

Discussion
This study was conducted to characterize the efficacy of soybean expressing Cry1Ac against H. gelotopoeon, S. albula, S. cosmiodes, S. eridania, and S. frugiperda and compare its performance with that of a non-Bt soybean with seed treatment by assessing defoliation (Figure 1). On the other hand, the performance analysis made among the species using non-Bt soybean without seed treatment showed which of these species would represent a potential risk for soybean crops. For both analyses, the defoliation and potential risk were calculated considering the identity for each species in greenhouse conditions (Table 1).
Our results showed that Bt soybean provided variable control of some pests evaluated. This new technology had the best control of H. gelotopoeon and a suppressive effect on S. frugiperda and S. albula.
Non-Bt soybean with seed treatment provided good control of the five species. Nevertheless, this control will be only for the early plant growth stage, when plants have more insecticide active ingredient in their system. S. cosmiodes, S. eridania, S. frugiperda, and S. albula feeding resulted in the lowest percentage of defoliation compared with the other treatments.
In the case of S. albula, the defoliation recorded in T1 was similar to that recorded in T2 and the differences were not significant.
These results are consistent with field studies evaluating Bt soybean in Brazil and Argentina. A study of diversity, composition, and population dynamics of arthropods in non-Bt soybean and Bt soybean showed that A. gemmatalis, C. includens, C. virescens, and S. frugiperda were significantly controlled by Bt soybean. However, other species of the Spodoptera complex were not controlled by this technology [70]. Other studies reported that S. cosmiodes was the most abundant species attacking Bt soybean [71] and that development and reproduction of this species were not affected by the Cry1Ac protein [20]. A field study showed that the lepidopteran pests recorded during the soybean cycle were R. nu, A. gemmatalis, H. gelotopoeon, S. frugiperda, and S. cosmioides [72] and the population abundance of the last two species was similar in Bt and non-Bt soybean. In northwestern Argentina, a trial conducted to evaluate Bt soybean against different insect pest and natural predators found that this new technology provided good control of target pests such as A. gemmatalis, C. includens, and R. nu. The levels of leaf damage observed with Bt soybean were lowest and significantly different to those obtained with non-Bt soybean [4]. On the other hand, bioassays and greenhouse studies found that S. cosmiodes, S. eridania, and S. frugiperda exhibited low to no susceptibility to MON 87701 × MON 89788 soybean containing the protein Cry1Ac, and these species showed higher tolerance to the Cry1Ac protein than other Lepidoptera species, such as C. includens, C. virescens, and H. zea [60]. In the same study, a moderate larval incidence of S. eridania and S. frugiperda on MON 87701 x MON 89788 soybean and the respective near-isogenic negative check was found. No significant differences in larval incidence and defoliation by S. eridania were found with Bt soybean and the near-isogenic negative checks for both maturity groups. Similarly, larval incidence of S. frugiperda on Bt soybean did not differ significantly from the respective near-isogenic negative checks for both maturity groups. However, defoliation by S. frugiperda on Bt soybean was significantly lower than on the near-isogenic, being the defoliation registered, similar to those reported in this study in Bt soybean. The defoliation recorded for S. eridania was different to those reported in greenhouse studies in Brazil [60], but in both studies, defoliation was high in non-Bt soybean. These differences could be due to the number of larvae used to evaluate the defoliation per plants.
The non-Bt soybean with seed treatment showed the lowest defoliation of S. frugiperda, S. albula, S. cosmiodes, and S. eridania. These results are similar to another study [73], where diamides had good potential to control different lepidopteran pests such as S. eridania, S. cosmiodes, S. albula, A. gemmatalis, and S. frugiperda.
Considering the performance of each species using non-Bt soybean without seed treatment, H. gelotopoeon showed the lowest defoliation. This species is one of the most important pests in soybean crops, but its larvae prefer the reproductive plant growth stages [50,74], which explains the low defoliation recorded in this study. As mentioned, Spodoptera spp. are polyphagous, but show preferences to different host plants, such as the preference of S. frugiperda to corn (Zea mays L. (Poaceae)) over soybean and other crops [75]. Spodoptera cosmiodes and S. albula presented similar levels of defoliation in our test, although S. cosmiodes is the most frequent Spodoptera spp. affecting soybean crops during the vegetative and reproductive stages [5,6,14,16,20,50]. Another study suggested that S. cosmioides may be more adapted to chemical compounds of soybean and cotton (Gossypium hirsutum L. (Malvaceae)), given their faster development and higher survival rates on these hosts [76]. Larvae of S. albula feed on a wide variety of host plants and they also exhibit some preference for several weeds (Boerhavia erecta (L.) (Nyctaginaceae); Echinochloa colonum (L.) Link (Poaceae), from which they can migrate to cultivated plants [27,29]. The highest defoliation damage was produced by S. eridania. This species had been cited as infesting a large number of crops in various regions of the Americas [49,74,77]. Additionally, this species has been reported from outbreaks under different conditions [15,18,28]. This species develops on weeds, which generally constitute a primary source of cultivated plant infestations such as those of soybean [15,26,28]. Considering the defoliation recorded in this study, S. eridania would represent a potential risk to soybean crops. According to these results, this species may be more adapted to chemical compounds of soybean given their faster development and higher survival rates [76].

Conclusions
Our results suggest that H. gelotopoeon, S. frugiperda, and S. albula were susceptible to the Cry1Ac protein. However, S. eridania and S. cosmiodes were not susceptible to the Cry1Ac protein in MON 87701 × MON 89788 soybean when evaluated by greenhouse infestation, supporting the idea that the level of activity of this protein against these species is low. Consequently, other control tactics, such as seed treatment, must be used in combination with MON 87701 × MON 89788 soybean in the field for the efficient management of Spodoptera species.
Our results contribute to determining the defoliation and potential risk of these pests in soybean crops. Considering the increasing importance of S. cosmioides and S. eridania in the region, future studies should be focused toward understanding survival, population dynamics, and infestation of these species during all growth stages of soybean plants, and their biology, including host adaptation. On the other hand, determination of the combined action of Bt with the seed treatment in the defoliation in the field will be important, like other control tactics. Therefore, the use of Bt soybean as a tool for integrated pest management should be planned according to the major insect problems in each area.